The aim of this project is to explore the utility of HLA and ABO tissue antigens as genetic and diagnostic markers for primary affective disorder (PAD) and its subtypes. The classification and genetic analysis of these disorders by clinical data alone are unsatisfactory. Progress will require the finding of specific biologic markers which associate with the state (and preferably with the trait) of PAD. Preliminary reports suggest that the HLA and ABO type frequencies differ between PAD and normal populations. In this study we will attempt to replicate the earlier reports and to extend them by correlating the results with a range of clinical and biologic variables. Using multivariate analyses we may be able to identify clusters of antigen types which associate with or dissociate from PAD (or other functional markers such as Xga blood group, color blindness, red blood cell lithium ratio, platelet MAO activity and response to specific treatments). The results may provide information which allows the use of antigen typing data for diagnostic and genetic purposes in PAD.